Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

An Elaborate Academic-Journal Hoax

chronicle.com

Post reply on HN
Print view View original

121–130 of 165 posts

Re: An Elaborate Academic-Journal Hoax

#121

Post by tomlock »

Sniffnoy wrote: post #117
tomlock wrote: post #84

Earlier quoted context omitted.

> and the "grievance scholars" are slowly convincing the rest of academia that real scholarship consists not of neutral methods and honest argument, but agreeing with the grievance scholars in all things, and finding ways to inject it whether it's relevant or not. What degrees or disciplines would you characterise as "grievance scholars", and what evidence do you have that suggests the influence those disciplines hav…

So, first off, "grievance studies" isn't my term. But as for evidence... well, I don't have statistics; it's more of a "look around you" thing. It's not clear to me why I would care about the number of degrees awarded in women's studies; that doesn't at all seem a reliable proxy for what I do care about, i.e., to what extent the academy is becoming less free and more ideological.

So, is it fair for me to say that your criticism of "grievance studies", a criticism of its unscientific and fallible nature, is anecdotal?

Re: An Elaborate Academic-Journal Hoax

#122

Post by tomlock »

darawk wrote: post #119
tomlock wrote: post #115

Earlier quoted context omitted.

All of them. Every field is fallible, phrenology was once just science. Why would you think something like that couldn't happen again?

You seem to be arguing that all epistemologies are fallible and therefore they are all equivalent. This is a pretty silly line of argument. The fact that a thing is fallible does not mean that it is the equal of things that don't even bother with the truth.

So, you feel like when someone says all fields are fallible and that you should examine their edification critically, that they are simultaneously making the claim that all fields are equivalent?

Do you feel you are fallible in your identification of which fields are concerned with the truth and which ones aren't?

Re: An Elaborate Academic-Journal Hoax

#123

Post by tomlock »

darawk wrote: post #112
tomlock wrote: post #91

Earlier quoted context omitted.

It's entirely possible to get a nonsense paper into an economics journal, even when not intentionally trying to expose the discipline as "nonsense". I'm not sure what you're not understanding here.

Replication crisis != nonsense papers, i'm not sure what you're not understanding here.

I guess I'm not understanding how you can think that a paper that makes a claim about the real world, but doesn't apply to the real world, could be characterised as anything but nonsensical.

Re: An Elaborate Academic-Journal Hoax

#124

Post by darawk »

tomlock wrote: post #122
darawk wrote: post #119

Earlier quoted context omitted.

You seem to be arguing that all epistemologies are fallible and therefore they are all equivalent. This is a pretty silly line of argument. The fact that a thing is fallible does not mean that it is the equal of things that don't even bother with the truth.

So, you feel like when someone says all fields are fallible and that you should examine their edification critically, that they are simultaneously making the claim that all fields are equivalent? Do you feel you are fallible in your identification of which fields are concerned with the truth and which ones aren't?

I'm saying that you have made no argument whatsoever that math and physics are equivalently fallible to gender studies, you've merely insinuated it. If you've got an argument to make, feel free to make it. But citing the replication crisis does not cut it.

Re: An Elaborate Academic-Journal Hoax

#125

Post by darawk »

tomlock wrote: post #123
darawk wrote: post #112

Earlier quoted context omitted.

Replication crisis != nonsense papers, i'm not sure what you're not understanding here.

I guess I'm not understanding how you can think that a paper that makes a claim about the real world, but doesn't apply to the real world, could be characterised as anything but nonsensical.

It makes an empirical claim using empirical data. Empirical data is complicated and inherently probabilistic. The fact that empirical studies sometimes conclude things that are wrong does not invalidate the notion of empirical investigation. It is an expected part of the distribution. It is also an issue that sometimes research practitioners do not follow the best practices, and this can lead to an exaggeration of false positive results beyond what you would expect from the underlying statistics. However, this also does not invalidate the idea of empirical investigation. It simply means we need to iterate on our training techniques and peer review.

What this paper asserts is not that gender studies need to iterate on their techniques. It does not assert that the basic methodology of gender studies is good, but it needs some tweaks here and there. It asserts that, at a fundamental level, the concept of the field is flawed. They assert that the entire methodological framework is flawed. No one, not the authors of the replication crisis paper, or Ioannidis, nor any other serious academic asserts this about math, physics, etc...

Re: An Elaborate Academic-Journal Hoax

#126

Post by int_19h »

foldr wrote: post #96
int_19h wrote: post #95

I see a lot of people complaining that it is done in bad faith, and doesn't prove much. But is there really much to prove? What's the baseline? I mean, we're talking about a field where this got published - and it was not a hoax: http://journals.sagepub.com/doi/full/10.1177/030913251562336... The people behind this hoax had an unenviable task of matching and beating a paper like that (since, apparently, by itself it…

Bad papers get published in all fields. Linking to one bad journal article shows nothing. Here, for example, is a reference to a bad paper published in a field which is not usually the target of this kind of criticism: https://academia.stackexchange.com/questions/9602/rediscover... https://math.berkeley.edu/~ehallman/math1B/TaisMethod.pdf

True, but when such papers are discovered, they're usually torn down by people in the field. In this case, it was the other way around - all meaningful criticism came from the outside, and in response, researchers in the field mostly circled the wagons.

Besides, the paper in question wasn't merely published - the author got a prestigious NSF award that lists that paper:

https://www.nsf.gov/awardsearch/showAward?AWD_ID=1253779

This, to me, implies that either such papers are considered broadly acceptable in the field, or they're not actually read. Not sure which is worse.

Re: An Elaborate Academic-Journal Hoax

#127

Post by tomlock »

darawk wrote: post #125
tomlock wrote: post #123

Earlier quoted context omitted.

I guess I'm not understanding how you can think that a paper that makes a claim about the real world, but doesn't apply to the real world, could be characterised as anything but nonsensical.

It makes an empirical claim using empirical data. Empirical data is complicated and inherently probabilistic. The fact that empirical studies sometimes conclude things that are wrong does not invalidate the notion of empirical investigation. It is an expected part of the distribution. It is also an issue that sometimes research practitioners do not follow the best practices, and this can lead to an exaggeration of fa…

Empirical data in individual studies are supposed to have statistical controls that mitigate incorrect conclusions, that's why p values are even a thing. To claim that on the one hand individual papers can be wrong, but the additive nature of a field somehow removes the possibility of error in the field, is misguided, to say it gently.

Could it be perhaps that this individual paper, which you claim undermines the "entire methodological framework" of these fields, could be one of those individual studies that comes to incorrect conclusions, not because of any statistical flaw, but because of the inbuilt fallibility and political biases of the authors?

Re: An Elaborate Academic-Journal Hoax

#128

Post by tomlock »

darawk wrote: post #124
tomlock wrote: post #122

Earlier quoted context omitted.

So, you feel like when someone says all fields are fallible and that you should examine their edification critically, that they are simultaneously making the claim that all fields are equivalent? Do you feel you are fallible in your identification of which fields are concerned with the truth and which ones aren't?

I'm saying that you have made no argument whatsoever that math and physics are equivalently fallible to gender studies, you've merely insinuated it. If you've got an argument to make, feel free to make it. But citing the replication crisis does not cut it.

Things are either fallible or they aren't, and since it seems you're willing to accept that all fields are fallible "to some degree" aka are fallible, we shouldn't edify any of them uncritically. Are you saying that you can claim things can be fallible without being equivalent? Because that's what I'm saying, and I'm glad we agree.

The replication crisis is a perfect example of the problem faced by all fields, of papers of questionable veracity being accepted by peer-reviewed journals. Convenient for you to say it is not up for discussion. However, I am casting a critical eye on your seemingly infallible nature, and I choose to consider it relevant.

Re: An Elaborate Academic-Journal Hoax

#129

Post by darawk »

tomlock wrote: post #128
darawk wrote: post #124

Earlier quoted context omitted.

I'm saying that you have made no argument whatsoever that math and physics are equivalently fallible to gender studies, you've merely insinuated it. If you've got an argument to make, feel free to make it. But citing the replication crisis does not cut it.

Things are either fallible or they aren't, and since it seems you're willing to accept that all fields are fallible "to some degree" aka are fallible, we shouldn't edify any of them uncritically. Are you saying that you can claim things can be fallible without being equivalent? Because that's what I'm saying, and I'm glad we agree. The replication crisis is a perfect example of the problem faced by all fields, of pap…

> The replication crisis is a perfect example of the problem faced by all fields, of papers of questionable veracity being accepted by peer-reviewed journals.

You're absolutely right. Now, feel free to explain how that is relevant to THIS issue.

Re: An Elaborate Academic-Journal Hoax

#130

Post by darawk »

tomlock wrote: post #127
darawk wrote: post #125

Earlier quoted context omitted.

It makes an empirical claim using empirical data. Empirical data is complicated and inherently probabilistic. The fact that empirical studies sometimes conclude things that are wrong does not invalidate the notion of empirical investigation. It is an expected part of the distribution. It is also an issue that sometimes research practitioners do not follow the best practices, and this can lead to an exaggeration of fa…

Empirical data in individual studies are supposed to have statistical controls that mitigate incorrect conclusions, that's why p values are even a thing. To claim that on the one hand individual papers can be wrong, but the additive nature of a field somehow removes the possibility of error in the field, is misguided, to say it gently. Could it be perhaps that this individual paper, which you claim undermines the "en…

Just pay attention to what i'm saying. I'm saying that the METHODOLOGICAL FRAMEWORK of physics and math is legitimate. The replication crisis et al does NOTHING to contradict that. This paper is ASSERTING that the METHODOLOGICAL FRAMEWORK of gender studies is flawed. You may disagree with that assertion. However, citing the replication crisis, which is a DIFFERENT CATEGORY OF ISSUE, does nothing to make your point.
Post reply on HN